Coarse PM Methods Evaluation Study · Coarse PM Methods Evaluation Study ... enters the APS inlet...

Coarse PMCoarse PMMethods Evaluation StudyMethods Evaluation Study

Study Design and Preliminary ResultsStudy Design and Preliminary Results

Thomas Ellestad, Robert Vanderpool, Thomas Ellestad, Robert Vanderpool, Paul Solomon, and Mary HarmonPaul Solomon, and Mary Harmon

USEPA/ORD/NERLUSEPA/ORD/NERL

Sanjay Natarajan, Chris Noble, and Bob MurdochSanjay Natarajan, Chris Noble, and Bob MurdochResearch Triangle InstituteResearch Triangle Institute

[email protected]@epa.gov919919--541541--78777877

BackgroundBackground

§§ 1997 air regulations established NAAQS for 1997 air regulations established NAAQS for PM2.5 and PM10 as separate metricsPM2.5 and PM10 as separate metrics

§§ U.S. courts have reviewed subsequent U.S. courts have reviewed subsequent litigation and ruled that the PM10 metric is a litigation and ruled that the PM10 metric is a “poorly matched indicator” because it “poorly matched indicator” because it includes the PM2.5 fractionincludes the PM2.5 fraction

§§ EPA has since been considering the possibility EPA has since been considering the possibility of vacating the PM10 regulation and of vacating the PM10 regulation and developing a separate standard for PMcdeveloping a separate standard for PMc

Study ObjectivesStudy Objectives

§§ EvaluateEvaluate the field performance of the field performance of leading leading methods for monitoring the coarse fraction of methods for monitoring the coarse fraction of PM10 (PMc = PM10 PM10 (PMc = PM10 –– PM2.5)PM2.5)

§§ Evaluate samplers which are either already Evaluate samplers which are either already commercially available or in their final stages of commercially available or in their final stages of developmentdevelopment

§§ Include both filterInclude both filter--based (timebased (time--integrated) and integrated) and semisemi--continuous measurement methodscontinuous measurement methods

PM2.5 and PM10 FRM SamplersPM2.5 and PM10 FRM Samplers§§ Standard lowStandard low--vol PM10 inlets vol PM10 inlets

aspirating at 16.7 lpm aspirating at 16.7 lpm (actual conditions)(actual conditions)

§§ PM2.5 aerosol fractionation PM2.5 aerosol fractionation using aWINS equipped with using aWINS equipped with DOS impaction oil DOS impaction oil

§§ Filters were conditioned at Filters were conditioned at 22C and 35% RH, analyzed 22C and 35% RH, analyzed gravimetrically. Postgravimetrically. Post--sampling filters archived at sampling filters archived at --30C for subsequent chemical 30C for subsequent chemical analysisanalysis

§§ 3 FRM pairs from BGI, 3 FRM pairs from BGI, R&P, and ThermoR&P, and Thermo--Andersen Andersen equipped with teflon filters equipped with teflon filters (4(4thth FRM pair equipped with FRM pair equipped with quartz filters)quartz filters)

PM10

PM10(WINS Removed)

PM2.5(with WINS)

PMc = PM10 – PM2.5

R&P PartisolR&P Partisol--Plus 2025 DichotPlus 2025 Dichot

§§ Standard PM10 inlet Standard PM10 inlet aspirating at 16.7 lpm aspirating at 16.7 lpm (actual)(actual)

§§ Aerosol fractionation by Aerosol fractionation by custom virtual impactor custom virtual impactor (15 lpm and 1.67 lpm)(15 lpm and 1.67 lpm)

§§ PM2.5 and PMc mass PM2.5 and PMc mass collected on 47 teflon filters collected on 47 teflon filters for gravimetric analysisfor gravimetric analysis

§§ Sequential sampler with Sequential sampler with multimulti--day capabilityday capability

§§ 4 units used in our study 4 units used in our study (3 teflon and 1 quartz)(3 teflon and 1 quartz)

R&P Coarse Particle TEOMR&P Coarse Particle TEOM

§§ Modified PM10 inlet Modified PM10 inlet aspirating at 50 lpm (actual)aspirating at 50 lpm (actual)

§§ PM10 aerosol is fractionated PM10 aerosol is fractionated by a custom virtual impactor by a custom virtual impactor (2 lpm coarse flow and 48 lpm (2 lpm coarse flow and 48 lpm fine flow)fine flow)

§§ PMc fraction is heated to 50 C PMc fraction is heated to 50 C to remove particle bound to remove particle bound waterwater

§§ Coarse aerosol is collected Coarse aerosol is collected and quantified by a standard and quantified by a standard TEOM sensorTEOM sensor

§§ 3 units used in our study3 units used in our study

Tisch SPMTisch SPM--613D Dichot Beta Gauge613D Dichot Beta Gauge

§§ Standard PM10 inlet Standard PM10 inlet aspirating at 16.7 lpm (~std)aspirating at 16.7 lpm (~std)

§§ Aerosol heated >25CAerosol heated >25C§§ Aerosol fractionation by Aerosol fractionation by

custom virtual impactor custom virtual impactor §§ PM2.5 and PMc mass PM2.5 and PMc mass

collected on polyflon tape collected on polyflon tape rollroll

§§ PM2.5 and PMc mass PM2.5 and PMc mass quantified hourly using quantified hourly using separate beta sources and separate beta sources and detectorsdetectors


TSI Model 3321 Aerodynamic Particle SizerTSI Model 3321 Aerodynamic Particle Sizer

§§ Standard PM10 inlet aspirating Standard PM10 inlet aspirating at 16.7 lpm (actual)at 16.7 lpm (actual)

§§ Isokinetic fraction of PM10 Isokinetic fraction of PM10 aerosol removed at 5 lpm and aerosol removed at 5 lpm and enters the APS inletenters the APS inlet

§§ APS sizes individual particles APS sizes individual particles aerodynamically using time of aerodynamically using time of flight approachflight approach

§§ Single particle volume converted Single particle volume converted to mass using mean density to mass using mean density provided by userprovided by user

§§ Total aerosol mass is sum of Total aerosol mass is sum of individual particle massindividual particle mass

§§ APS provide only PMc; not APS provide only PMc; not applicable for PM2.5 or PM10applicable for PM2.5 or PM10

§§ Only sampler in study which Only sampler in study which provides detailed size distribution provides detailed size distribution informationinformation


Mobile Sampling PlatformMobile Sampling Platform(Side View)(Side View)

Mobile Sampling PlatformMobile Sampling Platform(Front View)(Front View)

Sampler Performance IssuesSampler Performance Issues

§§ Relative bias versus collocated FRMs Relative bias versus collocated FRMs

§§ Precision (2 or 3 samplers of each type)Precision (2 or 3 samplers of each type)

§§ Field reliabilityField reliability

§§ Evaluation under a wide range of Evaluation under a wide range of weather conditions and aerosol typesweather conditions and aerosol types

QA/QC InitiativesQA/QC Initiatives§§ QAPP was reviewed and approved by EPAQAPP was reviewed and approved by EPA§§ Study design and operation passed EPA’s systems Study design and operation passed EPA’s systems

auditaudit§§ SOPs were reviewed by the sampler manufacturersSOPs were reviewed by the sampler manufacturers§§ Sampler manufacturers were allowed to verify the Sampler manufacturers were allowed to verify the

working condition of their respective samplers prior to working condition of their respective samplers prior to sampling at each sitesampling at each site§§ Sampling and fractionation components cleaned prior Sampling and fractionation components cleaned prior

to each studyto each study§§ NISTNIST--traceable sampler calibration equipment was traceable sampler calibration equipment was

used for all sampler calibrations and auditsused for all sampler calibrations and audits§§ Three performance audits and three field blank tests Three performance audits and three field blank tests

were conducted at each site were conducted at each site §§ Replicate weighings were conducted at the site as well Replicate weighings were conducted at the site as well

as at EPA’s RTP weighing facilityas at EPA’s RTP weighing facility

Study DetailsStudy Details

§§ Using 22 hour, daily sampling periods for comparisons Using 22 hour, daily sampling periods for comparisons (11 am to 9 am local time) (11 am to 9 am local time)

§§ Chemical analysis (XRF, IC, thermal optical) of Chemical analysis (XRF, IC, thermal optical) of archived filters will provide particle composition, archived filters will provide particle composition, which may explain observed sampler performancewhich may explain observed sampler performance

§§ Full data set from all three sites will be ready by Feb Full data set from all three sites will be ready by Feb 2004. Detailed results from Gary, IN will be presented 2004. Detailed results from Gary, IN will be presented at the October AAAR meetingat the October AAAR meeting

Study SitesStudy Sites

§§ RTP, NC (10 days of shakedown tests, Jan. 2003)RTP, NC (10 days of shakedown tests, Jan. 2003)

§§ Gary, IN (30 days of tests under cold, snow/rain, Gary, IN (30 days of tests under cold, snow/rain,

variable PM2.5/PM10 ratios, Marchvariable PM2.5/PM10 ratios, March--April, 2003)April, 2003)

§§ Phoenix, AZ (30 days of tests under hot, dusty Phoenix, AZ (30 days of tests under hot, dusty

conditions, consistently low PM2.5/PM10 ratios, conditions, consistently low PM2.5/PM10 ratios,

MayMay--June, 2003)June, 2003)

§§ Riverside, CA (30 days of tests under warm Riverside, CA (30 days of tests under warm

conditions, higher PM2.5/PM10 ratios than Phoenix, conditions, higher PM2.5/PM10 ratios than Phoenix,

JulyJuly--August, 2003)August, 2003)

Gary, INGary, IN

GARY, IN SIZE DISTRIBUTION DATAMarch - April, 2003

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Sample Day

PM

2.5/

PM

10 C

once

ntra

tion

Rat

io

GARY, IN FRM DATA

PM2.5/PM10 Range = 0.32 to 0.83; Mean = 0.55

PMc FRM MEASUREMENTS - GARY vs RTP WEIGHINGGary, IN (March - April, 2003)

0.0

10.0

20.0

30.0

40.0

50.0

60.0

0 5 10 15 20 25 30 35

SAMPLE DAY

PM

c C

on

c. (m

icro

gra

ms/

m3 )

GARY WEIGHING

RTP WEIGHING

RTP/Gary = 1.00

INTERMANUFACTURER PM2.5 FRM MEASUREMENTS (RTP WEIGHING)

Gary, IN (March - April, 2003)

0.0

10.0

20.0

30.0

40.0

50.0

60.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

SAMPLE DAY

PM

2.5

Co

nc.

(m

icro

gra

ms/

m3 )

ANDR&PBGI

INTERMANUFACTURER CV = 1.5%

Max PM2.5 = 46.9 ì g/m3

Min PM2.5 = 10.3 ì g/m3

Mean PM2.5 = 22.8 ì g/m3

INTERMANUFACTURER PM10 FRM MEASUREMENTS(RTP WEIGHING)


0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0

100.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

SAMPLE DAY

PM

10 C

on

c. (

mic

rog

ram

s/m

3 )

ANDR&PBGI


Max PM10 = 84.9 ì g/m3

Min PM10 = 22.6 ì g/m3

Mean PM10 = 42.6 ì g/m3

INTERMANUFACTURER PMc FRM MEASUREMENTS (RTP WEIGHING)


0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

SAMPLE DAY

PM

c C

onc.

(m

icro

gram

s/m

3 )

ANDR&PBGI


Max PMc = 58.1 ì g/m3

Min PMc = 4.5 ì g/m3

Mean PMc = 19.9 ì g/m3

DICHOT AND FRM TIMELINE (PM2.5)GARY, IN (MARCH - APRIL, 2003)

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

2.5 C

onc.

(mic

rogr

ams/

m3)

PM2.5 FRMR&P DICHOTS

DICHOT/FRM = 0.99

Dichot versus FRM PM2.5 ConcentrationsGary, IN (March - April, 2003)

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0

FRM PM2.5 Concentrations (micrograms/m3)

Dic

hot P

M2.

5 C

once

ntra

tions

(mic

rogr

ams/

m3)

Dichot = 0.99*FRM + 0.0R Square = 0.998

DICHOT AND FRM TIMELINE (PMc)Gary, IN (March - April, 2003)

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

c C

on

c.

(mic

rog

ram

s/m

3)

FRM PMc (PM10-PM2.5)R&P DICHOTS

Dichot/FRM = 0.89

Dichot = 0.87*FRM + 0.4R square = 0.969

DICHOT AND FRM TIMELINE (PM10)Gary, IN (March - April, 2003)

0

10

20

30

40

50

60

70

80

90

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

10 C

on

c. (m

icro

gra

ms/

m3 ) PM10 FRM

R&P DICHOTS

Dichot/FRM = 0.94Dichot = 0.95*FRM - 0.5R square = 0.981

TISCH SPM-613D AND FRM TIMELINE (PM2.5)Gary, IN (March - April, 2003)

0

10

20

30

40

50

60

70

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

2.5

Co

nc.

(mic

rog

ram

s/m

3 )

PM2.5 FRMTISCH SPM-613D

TISCH CV = 7.1%

TISCH/FRM = 1.26

Tisch = 1.17*FRM + 1.6R square = 0.948

TISCH SPM-613D AND FRM TIMELINE (PMc)Gary, IN (March - April, 2003)

0

10

20

30

40

50

60

70

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

c C

on

c. (m

icro

gra

ms/

m3 )

FRM PMc (PM10-PM2.5)

TISCH SPM-613D

TISCH/FRM = 0.91

TISCH = 0.89*FRM + 0.3R square = 0.979

TISCH CV = 10.5%

R&P COARSE TEOM AND FRM TIMELINE (PMc)Gary, IN (March - April, 2003)

0

10

20

30

40

50

60

70

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

c C

on

c. (

mic

rog

ram

s/m

3)

PMc (PM10-PM2.5)

R&P CONTINUOUS COARSE

TEOM PMc/FRM = 0.69

TEOM PMc CV = 4.4%

TEOM = 0.68* FRM + 0.18R square = 0.982

TSI APS vs FRM PMc ConcentrationsGary, IN (March - April)

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

SAMPLE DAY

PM

c C

on

c. (

mic

rog

ram

s/m

3 )

APS 1 PMc

APS 2 PMc

Mean FRM PMc

APS/FRM = 0.42

APS CV = 16.8%

APS = 0.48*FRM -1.4R square = 0.795

Phoenix versus RTP FRM WeighingMay - June 2003

0.0

50.0

100.0

150.0

200.0

250.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Sampling Day

Mas

s C

once

ntra

tion

(mic

rogr

ams

per

cubi

c m

eter

)

PM2.5 (RTP)

PM10 (RTP)

PMc (RTP)

PM2.5 (Site)

PM10 (Site)

PMc (Site)

PM10

PMc

PM2.5

Inter-manufacturerer PM2.5 CV = 3.4%Inter-manufacturerer PMc CV = 3.3%Inter-manufacturerer PM10 CV = 3.6%

PM2.5/PM10 Range = 0.10 to 0.28: Mean = 0.22

Dichot PM2.5 versus FRM PM2.5Phoenix, AZ

May - June, 2003

0.0

5.0

10.0

15.0

20.0

25.0

30.0

0.0 5.0 10.0 15.0 20.0 25.0

FRM PM2.5 (micrograms per cubic meter)

Dic

ho

t P

M2.

5 (m

icro

gra

ms

per

cu

bic

met

er)

Dichot = 1.24 * FRM - 1.6 R Square = 0.97

Mean Dichot/FRM = 1.09

DICHOT AND FRM TIMELINE (PMc)Phoenix, AZ (May - June, 2003)

0.0

50.0

100.0

150.0

200.0

250.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

c C

on

c.

(mic

rog

ram

s/m

3 ) PMc (PM10-PM2.5)R&P DICHOTS



DICHOT AND FRM TIMELINE (PM10)Phoenix, AZ (May - June, 2003)

0.0

50.0

100.0

150.0

200.0

250.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32

Sample Day

PM

10 C

on

c.

(mic

rog

ram

s/m

3 ) PM10 FRMR&P DICHOTS



Tisch & FRM PM2.5 ConcentrationsPhoenix AZ: May - Jun, 2003

0

10

20

30

40

50

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Sample Day

PM

2.5

con

cen

trat

ion

s (u

g/m

3)

Tisch PM2.5-1

Tisch PM2.5-2

Tisch PM2.5-3

Mean FRM-PM2.5

Average Tisch CV = 5.9 %

Mean Tisch/FRM = 1.70

Tisch = 2.03*FRM - 3.4R square = 0.947

Tisch, & FRM PMc ConcentrationsPhoenix AZ: May - Jun, 2003

0

20

40

60

80

100

120

140

160

180

200

220

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29

Sample Day

PM

c co

nce

ntr

atio

ns

(ug

/m3)

Tisch PMc-1

Tisch PMc-2

Tisch PMc-3

Mean FRM-PMc

Average Tisch PMc CV = 9.5 %

Mean Tisch/FRM = 1.04

Tisch = 0.92*FRM + 6.0R square = 0.996

TEOM, & FRM PMc ConcentrationsPhoenix AZ: May - Jun, 2003

0

20

40

60

80

100

120

140

160

180

200

220

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Sample Day

PM

c co

nce

ntr

atio

ns

(ug

/m3) TEOM PMc1

TEOM PMc2TEOM PMc3FRM-PMc

Average TEOM PMc CV = 6.6 %

Average TEOM/FRM Ratio = 1.05

TEOM = 0.79*FRM +12.8R square = 0.951

APS, & FRM PMc ConcentrationsPhoenix AZ: May - Jun, 2003

0

20

40

60

80

100

120

140

160

180

200

220

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Sample Day

PM

c co

nce

ntr

atio

ns

(ug

/m3)

APS 1-PMc

APS 2-PMc

FRM PMcAverage APS PMc CV (Days 1-15) = 2.2 %

Average APS/FRM = 0.55

APS = 0.56*FRM -0.2R square = 0.992

PM Size Distributions (TSI APS)Gary,IN and Phoenix,AZ

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0 1 10 100

Log Dp

Del

ta M

ass/

Del

ta L

og D

p

Gary

Phoenix

Summary of ResultsSummary of Results(independent of site)(independent of site)

§§ FRMs show strong interFRMs show strong inter--manufacturer precision manufacturer precision (CV<4% for all three metrics) with no tendency for (CV<4% for all three metrics) with no tendency for producing negative PMc values producing negative PMc values §§ FilterFilter--based dichots show strong precision (CV<4% based dichots show strong precision (CV<4%

for all metrics)for all metrics)§§ Site weighing results agree closely with RTP resultsSite weighing results agree closely with RTP results§§ Precision of the semiPrecision of the semi--continuous samplers is continuous samplers is

considered to be acceptableconsidered to be acceptable§§ Correlation (RCorrelation (R22) of all continuous samplers is ) of all continuous samplers is

typically strong versus the collocated FRMstypically strong versus the collocated FRMs

0.550.42APS/FRM PMcTSI APS

1.040.91Tisch/FRM PMc

1.161.09Tisch/FRM PM10

1.701.26Tisch/FRM PM2.5TISCH

1.050.69TEOM PMc/FRM TEOM PMc

0.790.90Dichot/FRM PMc

0.840.94Dichot/FRM PM10

1.091.00Dichot/FRM PM2.5DICHOTS

0.180.55PM2.5/PM10 Ratio

0.10 - 0.280.32 - 0.83PM2.5/PM10 Range

55.619.9PMc Mean ( ìg/m3)SITE AEROSOL

PHOENIX, AZGARY, IN

SUMMARY OF SITE RESULTSSUMMARY OF SITE RESULTS

Future WorkFuture Work

§§ Complete RTP gravimetric analysis of Riverside, Complete RTP gravimetric analysis of Riverside, CA filters (>1500 filter weighings per site)CA filters (>1500 filter weighings per site)

§§ Conduct chemical analysis of archived site Conduct chemical analysis of archived site filters; potentially use results as “explainers” of filters; potentially use results as “explainers” of sampler performancesampler performance

§§ Possibly conduct comprehensive field tests at an Possibly conduct comprehensive field tests at an additional field siteadditional field site

§§ Possibly perform laboratory tests with samplers Possibly perform laboratory tests with samplers to better understand aerosol fractionation and/or to better understand aerosol fractionation and/or particle loss issuesparticle loss issues

AcknowledgementsAcknowledgements

§§ RupprechtRupprecht & Patashnick, Inc.& Patashnick, Inc.§§ Tisch Environmental, Inc.Tisch Environmental, Inc.

§§ TSI, Inc.TSI, Inc.

§§ Indiana Department of Environmental Indiana Department of Environmental Management (Gary)Management (Gary)

§§ Maricopa County Environmental Services Maricopa County Environmental Services Department (Phoenix)Department (Phoenix)§§ University of California Ag Ops (Riverside)University of California Ag Ops (Riverside)

DisclaimerDisclaimer

§§ The United States Environmental Protection The United States Environmental Protection Agency through its Office of Research and Agency through its Office of Research and Development funded and managed the Development funded and managed the research described here under Contract 68research described here under Contract 68--DD--0000--206. It has been subjected to Agency 206. It has been subjected to Agency review and approved for publication.review and approved for publication.

Coarse PM Methods Evaluation Study · Coarse PM Methods Evaluation Study ... enters the APS inlet...

Documents

Transcript of Coarse PM Methods Evaluation Study · Coarse PM Methods Evaluation Study ... enters the APS inlet...